Search

Though in their infancy, mixed reality systems have proven to aid humans in mission critical tasks, once performed manually and requiring diverse skilled manpower. Industry 4.0 initiatives have seen mixed reality systems successfully adopted by many aerospace manufacturing environments, however aerospace maintenance and sustainment has been slower to adopt such initiatives, for various reasons. The paper reports about a mixed reality-based aircraft maintenance task that has been performed and compared with a non-mixed reality-based manual task. The data was collected using computer vision and machine learning techniques unobtrusively, generating both qualitative and quantitative measures, and a usability metric was derived. Furthermore, feature selection techniques are employed to determine the accuracy of the machine learning (ML) based usability assessment model. Supervised learning methods are employed for unstructured physiological data, and a usability metric for mixed-reality systems has been assessed based on dynamic features. The preliminary results are encouraging and lay the foundation for deriving data-driven models to predict usability of mixed reality systems when used in complex aircraft maintenance tasks. [ABSTRACT FROM AUTHOR]

The advent of electric propulsion in aviation comes with the opportunity to further increase propulsion system safety and reduce maintenance efforts as systems are becoming mechanically less complex. In modern electrical machines, the remaining components subject to mechanical wear are basically the rotor-bearings. Optimized designs use the same bearing for the rotor of the electrical machine and the propeller. One key technology to reduce maintenance cost and increase safety is to detect upcoming faults in a very early stage to enable condition-based maintenance and calculate remaining useful life (RUL). Whilst conventional and artificial neural network (ANN)-based monitoring techniques are well proven against available bearingfault-datasets in steady-conditions, bearing fault-detection and -classification becomes more complex in a non-steady environment. In this work, a 7-degree of freedom (DoF) bearing-fault model is used to synthesize high-resolution vibration data under different non-steady environments: Based on well validated 5-DoF simulation approaches, a bearing fault vibration model has been developed, allowing the synthesis of time-series signals for bearing-faults in a simulated (flight simulator) or recorded (flight experiments) environment. The model, implemented in MATLAB/Simulink, accounts for normal- and lateral-acceleration, pitch- and yaw-rates, propeller rotational-speed, and thrust-force to the bearing. By using simulated- and recorded-flight data, it can be shown that manoeuvring- and atmospheric-excitation has a significant impact on the bearing-fault vibration-signatures. The synthesized data can then be utilized to train ANN-based classifiers or regression-models, as well as to optimize detection algorithms based on conventional feature extraction for new propulsion-systems, before operational-data is available. Training and optimization of ANN-based classifiers is subject to ongoing work and not part of this paper. [ABSTRACT FROM AUTHOR]

Structural health management for Defence aerospace platforms is traditionally driven by scheduled inspection intervals and pre-emptive maintenance based on the interpretation of structural certification test results or reactive based on unique or unexpected fleet in-service incidents. However, a major shift towards proactive condition-based maintenance is required to improve availability and reduce fleet management and sustainment costs. Significant improvements to aerospace vehicle performance and structural health monitoring can be achieved by implementing an integrated approach to vehicle operation and sustainment. Numerous uncertainty factors and safety margins used in structural integrity assessment can be minimised or even eliminated by integrating a high-fidelity numerical model of the vehicle with a 'smart' health monitoring system and a comprehensive historical database for each individual vehicle covering its operation, maintenance, upgrades and repairs. Identifying and tracking the global airframe health as well as local structural anomalies caused by fatigue and wear is a crucial component towards the development of a smart structural diagnostics capability to support the operation and sustainment of the current and future aerospace platforms. This paper describes the progress of a collaborative effort between the Defence Science and Technology Group (DSTG) and RMIT University in developing a high-fidelity approach to structural health monitoring and individual aircraft tracking inspired by the Digital Twin paradigm. Operational Load Analysis and Asset Diagnostics (aka OPERAND) is a multi-physics analysis suite based on integrating current state-of-the-art software techniques, data-driven methods, and model-based approaches. This innovative structural diagnostic and prognostic framework has the potential to considerably improve fleet airworthiness management, provide substantial savings to fleet operators and optimise aerospace vehicle availability for enhanced operational effectiveness. [ABSTRACT FROM AUTHOR]

This paper describes a structural Prognostics and Health Management (PHM) framework for aircraft control surface free-play analysis. The data-driven diagnostic and prognostics approaches utilise on-board structural PHM data from accelerometers and an actuator load cell on an aircraft stabilator. The diagnostics and prognostics approaches include a two-phased process: i) detection of free-play severity via a supervised machine learning approach, useful for aircraft performance or near-term maintenance, and ii) estimation of Remaining Useful Life (RUL) with degradation modelling to support maintenance planning. Results demonstrate that a supervised machine-learning approach is capable of diagnostics from the time-series data with an overall accuracy of 86.4% using a lean classification decision tree model. The prognostics results show good prediction for a RUL estimation at early to mid-life, with a Root Mean Squared Error (RMSE) value of 37.47 samples and benchmark PHM score of 2.01×104, however, improvement can be made near the nominal failure threshold with an enhanced dataset and inclusion of operational events. The PHM framework serves to demonstrate a data-driven approach for aircraft maintainers to implement proactive condition-based maintenance as a supplementary analytical tool. Future work will incorporate more representative flight data and maintenance interventions, as well as increased complexity in diagnostic and prognostic models to handle non-linearities with further parametric fine-tuning. [ABSTRACT FROM AUTHOR]

This paper describes a Mixed Reality (MR) framework for structural inspections using an aircraft Digital Twin (DT) with Structural Health Monitoring (SHM) data. The purpose of this framework is to better support aircraft maintainers in performing inspection tasks and reduce the number of manual steps in unscheduled maintenance procedures. By combining DT and SHM data, this novel MR framework enables areas of potential fatigue damage to be visualised in 3-D on the full-scale aircraft. To contextualise the framework, a case study utilises a decommissioned Pilatus PC-9/A aircraft and operational loads monitoring data to demonstrate the MR approach. The fatigue damage 'hotspots' are calculated for six flights with varying severity of in-flight dynamic loads. The fatigue damage is mapped with the DT to generate the 3-D models of the 'hotspots' for use in an MR experience. Using a Head Mounted Display (HMD), such as the Microsoft HoloLens, the MR experience is viewed by the end-user when interacting with the actual aircraft. The MR approach aims to be a practical and supplementary tool for in-situ maintenance, visualising where fatigue damage 'hotspots' are located on the aircraft and facilitating compliance of inspections with interactive elements. Results show that the framework facilitates end-users to identify new information and could support fatigue damage maintenance inspections for the Pilatus PC-9/A. The MR experience developed shows a supplementary approach to identifying fatigue damage more proactively using conditionbased maintenance methods and over several flights identifying accumulated fatigue damage. [ABSTRACT FROM AUTHOR]